Swarm Optimized Feature Selection of EEG Signals for Brain-Computer Interface

A Brain-Computer Interface (BCI) is a communication system which uses cerebral activity to control external devices or computers. BCI research’s goal is to provide communication capability to the people who are totally paralyzed or suffer neurological neuromuscular disorders like amyotrophic lateral sclerosis, brain stem stroke or spinal cord injury. A BCI system records brain signals and applies machine learning algorithms to classify such signals, performing a computer controlled action. This study investigates effects of feature selection for Electroencephalograph (EEG) signals. Feature extractions using Walsh Hadamard Transform (WHT) and feature selection with Principal Component Analysis (PCA) are also studied. Feature selection through Particle Swarm Optimization (PSO) is proposed. Classification of the features is achieved through Bagging and decision tree classifiers.